[ RadSafe ] Low Dose Research

[howard long]

Bobby Scott has distributed this letter, perhaps of interest to many Radsafers.
"In conclusion, the biological data do not support LNT in the low-dose region."

Howard Long.

RADIATION RESEARCH 167, 742–744 (2007)
0033-7587/07 $15.00
_ 2007 by Radiation Research Society.
All rights of reproduction in any form reserved.
LETTERS TO THE EDITOR
Low-Dose Risk Assessment: Comments on the Summary of
the International Workshop
M. Tubiana,a A. Arengo,b D. Averbeckc and R. Massed
a Centre Antoine Be´cle`re, Faculte´ de Me´decine, 75006 Paris, France,
b Service Me´decine Nucle´aire, Hoˆpital Pitie´-Salpe´trie`re, 47 bd de
l’Hoˆpital, 75013 Paris, France, c Institut Curie-Section de Recherche,
Laboratoire Raymond Latarjet, UMR2027 du CNRS, Baˆtiment 110,
Centre Universitaire d’Orsay, F-91405 Orsay Cedex, France; and
d Acade´mie des Technologies, 75007 Paris, France
We read with much interest the summary of the international workshop
that was held at Columbia University. We were interested to learn that
David Brenner had the opportunity of explaining, during a lively discussion,
how and why the BEIR VII Committee and the French Academies
had come to diverging conclusions. We congratulate him because we
must admit that we still do not understand why the BEIR VII Committee,
while quoting most of the articles on which our conclusions are based,
did not mention many of them in its discussion and did not take into
account some of their data.
If one of the members of the French working party had the honor of
being invited to this meeting, he would have had the opportunity of explaining
our approach. First, we ask whether linear no-threshold (LNT)
is a postulate or a scientific hypothesis. If it is a scientific hypothesis,
this hypothesis should be clearly stated and its scientific bases analyzed
to facilitate the work of those who aim at confirming its validity or demonstrating
its falsehood. It seemed to us that the logic of proportionality
between dose and carcinogenic effect implies that the carcinogenic effectiveness,
per dose unit, remains constant irrespective of dose and dose
rate (1, 2). In the 1960s, when LNT was introduced, the proportionality
between dose and DNA damage was known but the existence of cell
defense mechanisms was largely unknown. Thus LNT was a plausible
hypothesis. Defense mechanisms have been much studied over the past
two decades. Their existence is not by itself an argument against LNT,
but proportionality between dose and carcinogenic or mutagenic effects
is plausible only if the effectiveness of defense mechanisms also remains
constant irrespective of dose and dose rate.
Let us look at the data. We know now that there are both specific
defense systems against some agents, such as solar ultraviolet irradiation,
and an array of protective mechanisms against a large variety of mutagenic
or carcinogenic agents. These protective systems are intertwined.
Furthermore, there are powerful signaling systems that rapidly inform a
cell about the insults that are caused to adjacent or distant cells. The
analysis that we made during the preparation of the report convinced us
that cell and tissue defenses present strong non-linearities (1, 2). The cell
defenses are modulated by the injuries that have occurred within the cell,
in the tissue or in other regions of the body, as was first shown several
decades ago. 

There are three main defense avenues against ionizing radiation,
such as at the cellular level:
1. Protection against reactive oxygen species by antioxidant molecules
(glutathione) and detoxifying enzymes such as catalase or SOD. These
mechanisms have a high effectiveness at low doses but a poorer effectiveness
at high doses. A low concentration of reactive oxygen
species triggers various defense mechanisms, and several genes are
involved in this response (1, 3).

2. DNA repair: DNA damage (such as DSBs) is detected by sensor molecules
that activate signaling factors that trigger cell cycle arrest and
DNA repair and apoptosis. Thus DNA repair functions are largely
controlled by DNA signaling: no repair if no signaling (1, 2, 4, 5).
The signaling systems (ATM and ATR) appear to be different at high
dose rate and low dose rate.
Several data show that the effectiveness of DNA repair is modulated
by dose and dose rate (4–9). During fractionated irradiation or lowdose-
rate irradiation, the death rate (per unit dose) is lower. This is
due to repair, as was shown by Elkind. The carcinogenic effect is also
lower (10), which also appears to be due to repair. Furthermore, the
influence of dose rate disappears when enzymatic repair systems are
blocked or absent. Vilenchik and Knudson (11) have shown that the
mutation rate also varies with dose rate. It has been shown that the
efficacy and fidelity of repair are higher when the amount of damage
present simultaneously in a cell is small (7). Furthermore, Joiner and
others have shown that there is a low-dose hypersensitivity that decreases
and disappears for doses _0.5 Gy (6, 8). This variation appears
to be linked to activation of repair systems since aminobenzamide
(an inhibitor of PARP) suppresses the decrease in sensitivity.
Hyperfast response also appears to be related to variations in the effectiveness
of DNA repair (1, 2). With regard to adaptive response,
the mechanisms involved are probably more complex, and membrane
damage (12) might play a role in the induction of cellular defense
systems in addition to DNA repair.

3. The third cell defense is elimination by death of cells with a damaged
genome. It is achieved through at least two mechanisms. The first is
apoptosis, which is an inducible mechanism. Its efficacy varies with
dose and becomes very low when the dose is greater than 300 mGy
(1, 2). It also varies widely with cell type and is, for example, much
greater for intestinal stem cells. This may explain the lack of second
cancers in the small intestine, a fact that confirms the impact of the
efficacy of apoptosis on the probability of carcinogenesis.
More recently, it has been shown that elimination of cells with
damaged genomes can also be due to lack of activation of DNA repair.
The experiments of Rothkamm and Lo¨brich (9) and Collis et al. (12)
show that at very low doses (_2 to 5 mGy) or dose rates, damaged
cells disappear rapidly. These data strongly suggest that this is due to
lack of signaling and therefore the absence of repair system activation.
Whatever the mechanism involved, the data clearly point out that at
low dose or dose rate, cells react differently to the insult caused by
ionizing radiation. Some scientists are skeptical about these data. After
a careful analysis and discussion with some of those who undertook
these experiments, the conclusion of our working party was that the
data mutually confirm each other and are valid. Furthermore, in a
recent paper, Lo¨brich et al. (13) confirm that technically at low and
very low doses their results are reliable. After a low dose (tomography)
of 20 mGy or a high dose of 2 Gy, damage was assessed with
comparable accuracy; the group was able to identify a deficiency in
DSB repair in cells derived from a patient showing individual hypersensitivity
to radiation therapy. However, the dose above which activation
of repair systems is triggered remains uncertain (around 5 mSv)
and probably varies with the tissue.
All these data lead to the view that at low dose (where few cells are
damaged) cell death is the simplest way to avoid the presence of initiated
cells, and this is the mechanism that has been selected during evolution
(1). When the dose is higher than about 200 mGy, the death of a large
number of cells may endanger the tissue; DNA repair becomes mandatory
despite the risk of misrepair. This conclusion is consistent with microarray
data showing that even very small doses (_2 mGy) provoke changes in
irradiated cells in the transcriptional expression of a large number of
genes, either activated or repressed (14, 15). The two important points
are that (1) the sets of genes that are activated or repressed are not the
same after a low dose and a high dose (14, 15) and (2) the time course
of the changes in gene expression are not the same after high doses (3 h
after 2 Gy) and low doses (only 48 h after 10 mGy) (15). More recently
(16), it has been shown that the effects of low and high doses on the
phosphoproteome differ not only quantitatively but also qualitatively.
These data confirm that it is illegitimate to extrapolate from high to low
doses (1, 2). These data also point out that the detection of an effect does
not mean that this effect is detrimental; it may very well be protective.
It has been said that at low doses the dose–effect relationship could be
supralinear due to bystander effects or genetic instability (17). With regard
to bystander effects, the recent results of Mothersill and colleagues
rule out this hypothesis. The dose threshold for bystander effects by medium
transfer is about 3 mGy (18). No reduction in survival is observed
for below 7 mSv. Bystanders may instead have a protective effect (19).
The absence of carcinogenic effects due to bystanders is in accordance
with the epidemiological surveys carried out on workers contaminated
with radium or patients contaminated with Thorotrast, among whom no
cancer excess is found when the cumulative dose is below a few grays
(1, 2). The bystander effect illustrates the interaction of cells through a
signaling network.
Could genetic instability induced by a low dose lead to carcinogenesis?
This hypothesis has not been supported by any experimental data (1).
Hence the carcinogenic effect of an irradiation (per unit dose) varies
strongly with dose and dose rate. These variations appear to be linked to
changes in the efficacy of cell defenses. Several data suggest the absence
of any detrimental effect for doses below 3 to 5 mSv. This is the dose
to which living cells have been exposed during evolution (natural background)
for 3.5 billion years. The existence of an effective defense in this
dose range is also not surprising since living organisms always develop
defense systems against toxic agents present in their external and internal
environments. It should be recalled that life appeared in the presence of
solar ultraviolet radiation and ionizing radiation.
The French Academies report has also discussed the tissue defenses
and the intervention of other cells during the carcinogenic process, in
particular the interaction between initiated cells, normal cells and the
stroma involving various cytokines (TGF_ and _) (1, 2, 20–22). The
mechanisms that govern embryogenesis and tissue repair also play a role
in the control of cell proliferation. Cancer is not simply a cellular disease
but involves dysfunction of the tissue control. The absence of cancer in
workers or patients contaminated by radium or Thorotrast when the cumulative
dose is below a few grays is probably explained by the fact that
the irradiated cells are surrounded by normal cells (1, 2). Clinical data
(for example regarding skin cancer or carcinoma of the cervix) suggest
that cell initiation is a frequent phenomenon and that the limitation of
the cancer incidence is related to the obstacles for a clone of transformed
cells to overcome defenses and to escape. Tissue disorganization caused
by the death of a large proportion of cells may facilitate the emergence
of a clone arising from a preneoplastic cell.
Defenses at the level of the organism (immunosurveillance) are another
effective defense mechanism, as shown by the high incidence of a few
types of cancer among immunodepressed individuals (patients with AIDS
or after a transplant) (23). High doses of radiation may lower immunological
defenses.

In conclusion, the biological data do not support LNT in the low-dose
region.

 Furthermore, epidemiology or experimental data, in vitro (24) or
in vivo (25, 26), provide evidence against a linear relationship. The main
deduction of LNT, namely that any dose, even the smallest, increases the
risk of cancer, does not take into account recent scientific data1 (17, 26).
Radiation protection was born prior to the LNT hypothesis and must
survive after LNT’s predictable death. Radiation protection is too important
to be linked with a highly debatable hypothesis. Finally, it should be
recalled that there is no epidemiological evidence of a carcinogenic effect
in humans for doses below 100 mSv (1, 2, 27).
How can we progress in this matter? What is needed is a dialogue, not
just talking but a true dialectical dispute with two aims: (1) To define the
biological implications of LNT on constancy in the effectiveness of cell
defense systems irrespective of the amount of DNA damage. A supralinear
dose–effect relationship means a lesser effectiveness when the
amount of damage is small and within the range of damage caused by
natural background. (2) To state unambiguously what are the points on
which proponents and opponents of the LNT hypothesis agree and disagree.
They should also describe the new experimental data or experiments
that could help one to accept or reject the LNT hypothesis.
The concept that any dose of radiation can increase the risk of cancer
is costly from both a financial and a human point of view. Every day it
raises anxiety among millions of individuals submitted to X-ray examinations.
Is this justified? The sooner we find an answer, the better.
Received: December 5, 2006
References
1. M. Tubiana, A. Aurengo, D. Averbeck, A. Bonnin, B. Le Guen, R.
Masse, R. Monier, A. J. Valleron and F. de Vathaire, Dose–effect
relationships and the estimation of the carcinogenic effects of low
doses of ionizing radiation. Acade´mie Nationale de Me´decine, Institut
de France–Acade´mie des Sciences, Paris, 2005. [English translation]
2. M. Tubiana, A. Aurengo, D. Averbeck and R. Masse, The debate on
the use of linear on threshold for assessing the effects of low doses.
J. Radiol. Prot. 26, 317–324 (2006).
3. L. E. Feinendegen and R. D. Neumann, Physics must join with biology
in better assessing risk from low dose irradiation. Radiat. Prot.
Dosimetry 117, 346–356 (2005).
4. D. Averbeck, L. Testard and D. Boucher, Changing views on ionizing
radiation-induced cellular effects. Int. J. Low Radiat. 3, 117–134
(2006).
5. D. Boucher, J. Hindo and D. Averbeck, Increased repair of gammainduced
DNA double-strand breaks at lower dose-rate in CHO cells.
Can. J. Physiol. Pharmacol. 82, 125–132 (2004).
6. A. Chalmers, P. Johnston, M. Woodcock, M. C. Joiner and B. Marples,
PARP-1, PARP-2, and the cellular response to low doses of
ionizing radiation. Int. J. Radiat. Oncol. Biol. Phys. 58, 410–419
(2004).
7. E. Dikomey and I. Brammer, Relationship between cellular radiosensitivity
and non-repaired double-strand breaks studied for different
growth states, dose rates and plating conditions in a normal fibroblast
line. Int. J. Radiat. Oncol. Biol. Phys. 76, 773–781 (2000).
8. B. Marples, B. G. Wouters, S. J. Collis, A. J. Chalmers and M. C.
Joiner, Low-dose hyper-radiosensitivity: A consequence of ineffective
cell cycle arrest of radiation-damaged G2-phase cells. Radiat.
Res. 161, 247–255 (2004).
9. K. Rothkamm and M. Lo¨brich, Evidence for a lack of DNA doublestrand
break repair in human cells exposed to very low x-ray doses.
Proc. Natl. Acad. Sci. USA 100, 5057–5062 (2003).
10. C. Rubino, F. de Vathaire, A. Shamsaldin and M. G. Leˆ, Radiation
dose, chemotherapy, hormonal treatment and risk of second cancer
after breast cancer treatment. Br. J. Cancer 89, 840–846 (2003).
11. M. M. Vilenchik and A. G. Knudson, Inverse radiation dose-rate
1 ICRP, Draft report of Committee I/Task Group, Low-dose Extrapolation
of Radiation-Related Cancer Risk, December 10, 2004.
744 LETTERS TO THE EDITOR
effects on somatic and germ-line mutations and DNA damage rates.
Proc. Natl. Acad. Sci. USA 97, 5381–5386 (2000).
12. S. J. Collis, J. M. Schwaninger, A. J. Ntambi, T. W. Keller, W. G.
Nelson, L. E. Dillehay and T. L. DeWeese, Evasion of early cellular
response mechanisms following low level radiation-induced DNA
damage. J. Biol. Chem. 279, 49624–49632 (2004).
13. M. Lo¨brich, N. Rief, M. Kuhne, J. Fleckenstein, C. Rube and M.
Uder, In vivo formation and repair of DNA double-strand breaks after
computed tomography examinations. Proc. Natl. Acad. Sci. USA 102,
8984–8989 (2005).
14. S. A. Amundson, R. A. Lee, C. A. Koch-Paiz, M. L. Bittner, P. Meltzer,
J. M. Trent and A. J. Fornace, Jr., Differential responses of stress
genes to low dose-rate gamma irradiation. Mol. Cancer Res. 1, 445–
452 (2003).
15. N. Franco, J. Lamartine, V. Frouin, P. Le Minter, C. Petat, J-J. Leplat,
F. Liber, X. Gidrol and M. C. Martin, Low-dose exposure to _ rays
induces specific gene regulations in normal human keratinocytes. Radiat.
Res. 163, 623–635 (2005).
16. F. Yang, D. L. Stenoien, E. F. Strittmatter, J. Wang, L. Ding, M. S.
Lipton, M. E. Monroe, C. D. Nicora, M. A. Gristenko and R. D.
Smith, Phosphoproteome profiling of human skin fibroblast cells in
response to low- and high-dose irradiation. J. Proteome Res. 5, 1252–
1260 (2006).
17. D. J. Brenner, R. Doll, D. T. Goodhead, E. J. Hall, C. E. Land, J. B.
Little, J. H. Lubin, D. L. Preston, J. S. Puskin and M. Zaider, Cancer
risk attributable to low doses of ionizing radiation: Assessing what
we really know. Proc. Natl. Acad. Sci. USA 100, 13761–13766
(2003).
18. Z. Liu, C. E. Mothersill, F. E. McNeill, F. M. Lyng, S. H. Byun, C.
B. Seymour and W. V. Prestwich, A dose threshold for a medium
transfer bystander effect for a human skin cell line. Radiat. Res. 166,
19–23 (2006).
19. O. V. Belyakov, M. Folkard and C. Mothersill, Bystander induced
differentiation. A major response to targeted irradiation of a urothelial
explant model. Mutat. Res. 597, 43–49 (2006).
20. M. M. Mueller and N. E. Fusening, Friends or foes. Bipolar effects
of the tumour stroma in cancer. Nat. Rev. 4, 839–849 (2004).
21. D. C. Radisky and M. J. Bissell, Cancer. Respect thy neighbor! Science
303, 774–775 (2004).
22. M. H. Barcellos-Hoff, Integrative radiation carcinogenesis: interactions
between cell and tissue responses to DNA damage. Semin. Cancer
Biol. 15, 138–148 (2005).
23. S. Euvrard, J. Kanitakis and D. Claudis, Skin cancers after organ
transplantation. N. Engl. J. Med. 348, 1681–1691 (2003).
24. M. Ko, X. Y. Lao, R. Kapadia, E. Elmore and J. L. Redpath, Neoplastic
transformation in vitro by low doses of ionizing radiation:
Role of adaptive response and bystander effects. Mutat. Res. 597,
11–17 (2006).
25. H. Tanooka, Threshold dose-response in radiation carcinogenesis: An
approach from chronic beta-irradiation experiments and a review of
non tumour doses. Int. J. Radiat. Oncol. Biol. Phys. 77, 541–551
(2001).
26. P. Duport, A database of cancer induction by low dose radiation in
mammals: Overview and initial observations. Int. J. Low Radiat. 1,
120–131 (2003).
27. J. Breckow, Linear-no-threshold is a radiation-protection standard
rather than a mechanistic effect model. Radiat. Environ. Biophys. 44,
257–260 (2006).